Update StationaryObject::Classification::Type with new field for park…

[yonivai]

#797

[thomassedlmayer]

This was already discussed some time ago (see #714) but we did not come to any conclusion because the drivers of the discussion dropped out or lost interest. However, as Lorenz said, having free space marked as object in OSI may introduce problems. @PhRosenberger @yonivai Did you consider this in your discussion?
What is the use case for this?

[PhRosenberger]

@PhRosenberger @yonivai Did you consider this in your discussion? What is the use case for this?

I was not aware of #714, however the use case is to actually detect a parking spot as a bounding box, so this is something like a stationary object to me, even if it is a special case, as you can actually drive into it. On the other hand, there are stationary objects like a newspaper on the street that are kind of similar. This is the reason, why I would vote for adding this type to OSI.

[yonivai]

This was already discussed some time ago (see #714) but we did not come to any conclusion because the drivers of the discussion dropped out or lost interest. However, as Lorenz said, having free space marked as object in OSI may introduce problems. @PhRosenberger @yonivai Did you consider this in your discussion? What is the use case for this?

Hi @thomassedlmayer,
as @PhRosenberger said, we weren't aware of this discussion.
I do think that Ground Truth wise there is place for a parking spot to be represented.
if we are looking at a used case scenario, we are seeing it as a static representation of the world (like a map characteristics , before the scenario begins they already exists, and will be for the rest of the scenario).
also, we already have some types that a vehicle can drive onto it in the types of the stationary object:
https://opensimulationinterface.github.io/open-simulation-interface/structosi3_1_1StationaryObject_1_1Classification.html#ad497488a7119617633b54181ea08457da0035009db5b2fadfc746e432cd4d3eb9
https://opensimulationinterface.github.io/open-simulation-interface/structosi3_1_1StationaryObject_1_1Classification.html#ad497488a7119617633b54181ea08457da3061d9cf215369f5d91dad750d08e05d
https://opensimulationinterface.github.io/open-simulation-interface/structosi3_1_1StationaryObject_1_1Classification.html#ad497488a7119617633b54181ea08457daebee7560a0a5b63846221860fcf64321

[yonivai]

Hi, is there any news on this matter?
will we progress on that? or should we discuss about it?

[thomassedlmayer]

As the last meeting of the responsible working group took place before this PR was opened, this unfortunately won't make it into the release. The review phase for the 3.7.0 release has started today.
But we may pursue this issue in future harmonization efforts. There is a working group developing non-normative harmonization specifications for various ASAM standards (e.g., OpenDrive, OpenScenario, OSI). This group is currently looking at traffic participant harmonization, but the aim would be to take forward other issues such as parking space representation in the future. I think it makes sense to check if the representation of a parking space as stationary object is still appropriate in other standards before adapting it.

[PhRosenberger]

I would like to add something to @thomassedlmayer's last comment:
As mentioned, there are several working groups to consider for this topic.
Harmonization is important and I have been informed that other standards groups such as OpenDRIVE are also discussing related issues with respect to parking use cases.

@yonivai we would propose to use this internally as already suggested, as we also see a good chance to integrate it as suggested by you.
However, it was not possible to include it in the upcoming release due to the already started review phase. So we would encourage to participate in the next project phase towards the next release of OSI.